The Agent Skills Directory

[DATA_EXFILTRATION] (LOW): The skill utilizes curl in Pattern 3 to interact with an external API_BASE_URL. While this is standard for API testing, it involves network operations to non-whitelisted domains.
[COMMAND_EXECUTION] (SAFE): Use of run and bash -c is appropriate for the context of a Bash testing framework. No signs of arbitrary or malicious command execution were found.
[PROMPT_INJECTION] (LOW): The skill processes untrusted data from API responses, creating a surface for indirect prompt injection (Category 8).
Ingestion points: External API data is captured via curl into HTTP_BODY and $body_file in Pattern 3.
Boundary markers: No explicit boundary markers or instructions to ignore embedded commands are present in the provided templates.
Capability inventory: The agent uses run for subprocess execution, curl for network requests, and jq for data processing.
Sanitization: The skill recommends using jq for JSON assertions, which provides structural validation but not full sanitization against instructional content.

bats-testing